The three most important research needs for progress in the treatment of tuberculosis are to (a) develop drugs against drug resistant strains of Mycobacterium tuberculosis, (b) develop effective approaches to treat the latent states of the disease, and (c) shorten the course of therapy, which is related to the second goal. Cholesterol is crucial for infection of macrophages and survival of the mycobacteria in that environment. In the expiring period of support, we have identified three M. tuberculosis cytochrome P450 enzymes able to initiate cholesterol catabolism. Our studies have identified an intermediate, cholest-4-en-3-one that accumulates when these P450 enzymes are knocked out. This intermediate inhibits growth of MTB on several carbon sources. We propose to identify the site and mechanism of action of cholest-4-en-3-one, as it is a potential drug target. Furthermore, we will undertake the development of mechanism-based inhibitors of the two primary enzymes involved in degradation of the cholesterol side-chain, as their inactivation will not only block cholesterol utilization but will lead to accumulation of cholest-4-en-3-one. In a second facet of this project, we will characterize and define the biological roles of two P450 enzymes related to virulence that are proposed to oxidize methyl-branched hydrocarbon chains. In a third facet of our P450 studies, we will advance our work to define the structures, substrates, and roles of the other M. tuberculosis P450 enzymes. Of the twenty P450 enzymes, we now know the substrates and functions of four. We have cloned the remaining 16, partially characterized four, and have the structure of one. The collective evidence indicates that the P450 enzymes in MTB have specific biological roles rather than being involved in xenobiotic metabolism. In a related but distinct effort, we will further characterize the redundant DosS/DosT/DosR two-component regulatory systems of M. tuberculosis. These sensors control a regulon of approximately 50 genes that is induced by hypoxia, NO, or CO. The associated metabolic shift is thought to be similar to that which initiates the dormant state of M. tuberculosis. We have demonstrated that the sensors are heme proteins, defined the differential response to the various gases, and determined the crystal structure of DosT. We now propose to complete definition of the system by analyzing the mechanism by which the identity of the gas binding to the heme iron atom is transferred to the kinase domain of the sensor as either an off- (O2) or on- (NO, CO) signal. We will also perform a high-throughput search for inhibitors of the DosS / DosT kinases that may provide lead compounds for the development of agents effective against the latent states of M. tuberculosis. The proposed work rests on cutting-edge techniques, including proteomic, lipidomic, and metabolomic analysis of M. tuberculosis mutants and knockouts, NMR studies of protein conformational changes with 13C- and 19F-labeled site-specifically incorporated amino acids, and chemical biological studies of the roles of steroids and other factors in the biology of M. tuberculosis.